Diabetes Mellitus is a chronic condition characterised by a metabolic abnormality in which the pancreas fails to make enough insulin or cells fail to respond to the insulin produced, resulting in excessive blood sugar levels. Finding this disease at the early stages reduces the risk of patients from having more complicated health problems. The main reason considered for the constant increase in this disease is the lack of awareness of the importance of healthy eating habits. According to recent predictions of rising diabetes, the global population of diabetic patients will reach 642 million in 2040, making diabetes a problem for one out of every 10 persons. Machine learning techniques have been applied to various sectors of medical health due to the rapid development of the subject. In this work, Considering the importance of early medical diagnosis of diabetes disease, many efficient machine learning algorithms are proposed for the identification of type Diabetes Mellitus in patients. Different machine learning models, such as support vector machines, Random Forest classifier, and K-Nearest Neighbour classifiershave been applied and compared to study the accuracy of the algorithms.

Classification of Diabetes Mellitus, Support Vector Machine, Grey Wolf Optimization Treatments